We're just quibbling over terminology, but the point over which we're quibbling is important. To say: is to express a hypothesis, not an assumption, because of the conditional word "may." You're not saying, "I have an unshakeable faith that, regardless of the existence or absence of evidence, there are invisible pellets in this tube." That's an assumption, and that's precisely the way nearly all religionists "believe" in their mythologies. What you're saying is, "Based upon my experience with this apparatus and the past behavior of the universe (i.e., under these conditions there has always been a pellet), I judge there to be a reasonable probability that there is a pellet here, and due to unusual but not extraordinary conditions, it's obscured from view." That is a hypothesis. Like all hypotheses, yours was based on both evidence and reasoning, and you were not certain that it was true, so you went on to test it like a good scientist. An assumption may be based on evidence and reasoning too, in which case the distinction from a hypothesis is pedantic, but it may also be based on creative or wishful thinking, or on a preprogrammed instinctive belief. I suppose an assumption can be tested by the scientific method, in which case it's treated as a hypothesis. To once again remind ourselves that this is the Linguistics Board, we should note that perhaps the definition of "hypothesis" can include "assumption," but it's key to the practice of science that our laboratories not be overbooked by people testing assumptions that are no more than crackpottery. This might even be a corollary to Occam's Razor ("Test the simplest explanation for a phenomenon first and get it out of the way before you take a chance on wasting your time testing a complicated explanation"). An explanation for a phenomenon that is not based on evidence and reasoning is unlikely to be a simple explanation to test, and it could easily be a colossal waste of time Of course. But you're not assuming it's there, you're simply reasoning, based on past evidence, that it might be there. That's a big difference. Sure, but it's a reasoned faith, not an irrational, instinctive faith. That makes all the difference. Your faith in the mere possibility of an invisible pellet was reasoned. You apparently did have a reason for thinking that, because the first thing you did was wash the tube. You reasoned that something could be obscuring your view. You didn't take it to an exorcist. Why is that an assumption? We have been performing the act that we call "observing" for millions of years, since long before we had a forebrain powerful enough to philosophize about it, and that act has gathered data about the natural universe that we have tested for all those millions of years and found to be acceptably reliable. It's not even a hypothesis any more. It's a canonical theory that transcends science. We now know that there are exceptions such as intoxication, illness, hypnotism, brainwashing, and in court trials we have even learned that what people remember about what they see can be affected by what they expected to see, but nonetheless observation has been "proven beyond a reasonable doubt" to be satisfactory evidence in the practice of the scientific method, so long as it is corroborated by testing and peer review. I think I just covered that. This is a textbook explanation of why scientific experiments must be duplicated during a peer review. Excuse me??? Myths purport to explain things, but they tend to operate at a metaphorical level, at best. Many of them, as I've explained a number of times, are instincts that Jung calls archetypes, motifs preprogrammed into our synapses by evolution. Some are survival traits from an era whose dangers we can't imagine, and others are the random result of genetic drift and bottlenecks. Give us a for-instance. And please pick something reasonably current, not from the 16th century when we were still getting our scientific act together. That's hardly the purpose of science. To the extent that it works out that way, it's merely the inevitable result of the fact that so many of our myths are preprogrammed archetypes, and conflict with reality. This is where metaphor comes in, but a surprising percentage of the population have no place in their cognitive process for metaphor. If there's something regrettable about the human condition that we've identified in this discussion, that might very well be it. To tie this up with the topic of this thread: Religion is metaphor.